+ All Categories
Home > Documents > Designing a Chipseal Agg and Binder 2014 NWPMA 2014 · Discussion)Topics • Mcleod)Chipseal ......

Designing a Chipseal Agg and Binder 2014 NWPMA 2014 · Discussion)Topics • Mcleod)Chipseal ......

Date post: 27-May-2018
Category:
Upload: phungkhuong
View: 214 times
Download: 0 times
Share this document with a friend
57
Chipseal Design and Materials Stephen Van De Bogert Western States Asphalt
Transcript

Chipseal  Design  and  Materials

Stephen  Van  De  Bogert  Western  States  Asphalt

Discussion  Topics

• Mcleod  Chipseal  Design  (  Mndot  version)  • Aggregate  Requirements  • Binders  • Alternative  to  a  Chipseal  

– Maintenance  Seal  

• Review  of  Spokane’s  FA-­‐2  Seals  

Chipseal  Design  Method

• What  should  this  design  method  do?  – 1.    Give  amount  of  aggregate  needed  to  cover  1  sq.  yd²  a  single  stone  thick  

– 2.    Give  starting  binder  application  rate  • Starting  rate  would  yield  60%  to  70%  embedment  if  no  absorption  by  pavement  

• Must  adjust  for  current  conditions  of  pavement  – Recommendation  for  crew  to  use  to  help  adjust  for  traffic  and  conditions  of  pavement

Design  

• Design  is  based  upon  a  single  rock  source  /  sample  – Each  rock  source  needs  a  design    Do  not  assume  two  sources  meeting  the  same  spec  are  close  enough.  

• Takes  into  account  traffic  effects  – The  higher  the  traffic  the  more  compaction  the  surface  gets  and  the  lower  the  binder  content  to  hold  the  rock  and  vice  versa.  

• Takes  into  account  road  surface  conditions  – The  rougher  the  road  the  more  binder  it  going  to  be  absorbed  into  the  surface  so  more  binder  is  needed  to  have  enough  left  to  hold  the  chips.

• Chip  Seal  Design  Program    http://www.dot.state.mn.us/materials/researchsealcoat.html  

• Minnesota  Seal  Coat  Handbook  http://www.lrrb.org/pdf/200634.pdf

Reference  Source

McLeod  Emulsion  Calculation MNDOT  Version

• B(G/SqYd)=((2.244  x  Ave  Least  Dimension  x  Traffic  Factor  x  Voids  in  loose  Agg)  +  Surface  Condition  factor  +  Agg  Absorb  Factor)  /  Residual  Asphalt  Content  of  Binder  .    For  Wheel  Paths  

• Then  same  calc  on  Median  Rock  Size  instead  of  Average  Least  Dimension.    For  Non  Wheel  Paths  

• Average  the  two.  • Only  need  calc  on  Median  Rock  Size  if  rock  is  very  cubical.      

• May  not  need  to  average  if  we  can  improve  non  -­‐wheel  path  embedment.    

Precision  is  the  Key  to  Success

• The  higher  the  number  of  sieves  used  to  grade  the  material  the  more  accurate  the  design.  

• The  more  cubical  the  rock  the  more  precise  the  design.  

• Accurate  traffic  count.    • Accurate  Road  evaluation.

Sieve Passing1/2 0.5 1003/8 0.375 964 0.187 910 0.0787 3200 0.0029 1.3

0

10

20

30

40

50

60

70

80

90

100

110

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55

%+Passin

g

Sieve

Sieve Passing1/2 0.5 1003/8 0.375 961/4 0.25 204 0.187 910 0.0787 3200 0.0029 1.3

0

10

20

30

40

50

60

70

80

90

100

110

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55

%+Passin

g

Sieve

0

10

20

30

40

50

60

70

80

90

100

110

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55

%+Passin

g

Sieve

Tests  run  on  Aggregate  for  design  

• Gradation  –  Binder  ;  for  embedment  • Loose  Unit  Weight  –  To  Calc.  Voids  /  room  for  binder  

• Specific  Gravity  –    To  Calc.  Voids    • Absorption  –  Binder;  for  loss  in  absorb.  • Flakiness  Index-­‐  Binder;  for  functional  embedment.    How  high  will  the  chips  sit  up  when  finally  embedded.  

H Ave Least Dimension 0.294 Flakiness ratio 0.9 0.294

T Traffic Factor -100 ADT 0.85 100-500 0.75 500-1000 0.7 1000-2000 0.65V Voids in loose Agg Single Chip 0.5S Surface

ConditionSmooth, non porous

0 Slightly porous & oxidized

0.03 Slightly pocked porous & oxidized

0.06 Badly pocked porous & oxidized

0.09

A Agg Absorption None 0R Residual AC Cont 0.665

-100 ADT on Badly Pocked & oxidized

Binder Appl Rate = 0.557 Gal/yd2 2.244 0.294 0.85 0.5 0.09 00.665

-100 ADT on Slightly pocked, Porous & oxidized

Binder Appl Rate = 0.512 Gal/yd2 2.244 0.294 0.85 0.5 0.06 00.665

-100 ADT on Slightly Porous & oxidized

Binder Appl Rate = 0.467 Gal/yd2 2.244 0.294 0.85 0.5 0.03 00.665

-100 ADT on Smooth non-porous

Binder Appl Rate = 0.422 Gal/yd2 2.244 0.294 0.85 0.5 0 00.665

200-500 ADT on Badly Pocked & oxidized

Binder Appl Rate = 0.507 Gal/yd2 2.244 0.294 0.75 0.5 0.09 00.665

200-500 ADT on Slightly Pocked, Porous & oxidized

Binder Appl Rate = 0.462 Gal/yd2 2.244 0.294 0.75 0.5 0.06 00.665

Project Name:

Median Rock Size

This Design is for reference only. Field adjustments are necessary. Design done from 1 sample supplied by the customer and may vary due to stockpile variations, errors in sampling etc.

Flora Pit 2011

Project Name: Flora Pit 2011

This Design is for reference only. Field adjustments are necessary. Design done from 1 sample supplied by the customer and may vary due to stockpile variations, errors in sampling etc.

H Ave Least Dimension Screen with 1st. 20% Retained

0.294 Flakiness ratio 0.8 0.2646

T Traffic Factor -100 ADT 0.85 100-500 0.75 500-1000 0.7 1000-2000 0.65V Voids in loose Agg Single Chip 0.5

S Surface Condition

Smooth, non porous

0 Slightly porous & oxidized

0.03 Slightly pocked porous & oxidized

0.06 Badly pocked porous & oxidized

0.09

A Agg Absorption None 0R Residual AC Cont 0.665

-100 ADT on Badly Pocked & oxidized

Binder Appl Rate = 0.515 Gal/yd2 2.244 0.2646 0.85 0.5 0.09 0 0.665

-100 ADT on Slightly pocked, Porous & oxidized

Binder Appl Rate = 0.470 Gal/yd2 2.244 0.2646 0.85 0.5 0.06 0 0.665

-100 ADT on Slightly Porous & oxidized

Binder Appl Rate = 0.425 Gal/yd2 2.244 0.2646 0.85 0.5 0.03 0 0.665

Let’s  make  it  a  less  cubical

Project Name: Flora Pit 2011

This Design is for reference only. Field adjustments are necessary. Design done from 1 sample supplied by the customer and may vary due to stockpile variations, errors in sampling etc.

H Ave Least Dimension Screen with 1st. 20% Retained

0.294 Flakiness ratio 0.9 0.294

T Traffic Factor -100 ADT 0.85 100-500 0.75 500-1000 0.7 1000-2000 0.65V Voids in loose Agg Single Chip 0.5

S Surface Condition

Smooth, non porous

0 Slightly porous & oxidized

0.03 Slightly pocked porous & oxidized

0.06 Badly pocked porous & oxidized

0.09

A Agg Absorption None 0.02R Residual AC Cont 0.665

-100 ADT on Badly Pocked & oxidized

Binder Appl Rate = 0.587 Gal/yd2 2.244 0.294 0.85 0.5 0.09 0.02 0.665

-100 ADT on Slightly pocked, Porous & oxidized

Binder Appl Rate = 0.542 Gal/yd2 2.244 0.294 0.85 0.5 0.06 0.02 0.665

-100 ADT on Slightly Porous & oxidized

Binder Appl Rate = 0.497 Gal/yd2 2.244 0.294 0.85 0.5 0.03 0.02 0.665

Now  higher  Agg.  Absorption

Ba00-0222 for TH 44

BIN

DER

APP

LIC

ATIO

N R

ATE

(gal

/yd2

)

0.26

0.46

TRAFFIC VOLUME (ADT)Over 2,000 1,000 to 2,000 500 to 1,000 100 to 500 Under 100

Badly pocked, porous, and oxidized

Slightly pocked, porous, and oxidized

Slightly porous, and oxidized

Aggregates

Standard  Chipseal  Aggregate  Requirements

• Must  be  clean  -­‐  For  reliability  – Less  then  1-­‐2%  passing  #200  sieve;  better  adhesion  

• Durable  -­‐  wear  life  – LAR,  lower  =  harder,  polish    /  wear    resistant  

• Flakiness  Index  -­‐  reliability  – Lower  =  More  cubicle,  uniform  shape    easier  to  design  around.    More  accurate  design  =  More  reliable  seal.  

• Need  to  have  fractured  faces  for  stability

Aggregate  Application  Rate  Calc• Need    -­‐  Ave  least  Dimension  • Need    -­‐  Specific  Gravity  of  the  Aggregate  • Calculate  the  Voids  in  Loose  Agg  

– V=  (Loose  unit  weight  (lbs/cubic  ft)/(62.4*Spec  Gravity)  

• Wastage  factor  Example  10%  for  high  traffic,  5%  for  very  low  slow  traffic    1+.10  +1.1  high  Traffic  

• C  (Appl  Rate)  =  46.8*  (1-­‐(0.4)*Voids  in  loose  Agg*Ave  Least  Dimension*Specific  Gravity  *Wastage  Factor  for  Traffic  whip  off

– Single sized Chips  • More uniformed height  • Has more room for binder – Space not filled by smaller

aggregate particles.  • The more single sized the easier it is to develop a good

chipseal design.

The  Problem  with  Flat  Chips

Binders

CRS-­‐2P,  CMS-­‐2P

• Polymer  Emulsions  • Stiffer  binder    -­‐  reduces  bleeding  • Develops  strength  faster  than  other  emulsions,  can  sweep  sooner.  

• Requires  clean  chips  • Must  place  chips  immediately    • Most  Expensive  conventional  chipseal  emulsion

Specification  Change  to  CRS-­‐2P  

• Beginning  with  2015  specification  change  to  the  CRS-­‐2P  specification  

• AASHTO  T  301  Elastic  Recovery  Test    -­‐  50%  min  will  replace:    

• Torsional  Recovery    -­‐  18%  min            And  the  alternate  • Toughness  and  Tenacity  test  -­‐  50/25  min.

CMS-­‐2P

• There  is  no  WSDOT  specification  for  this  material.  

• Generally  very  close  to  CRS-­‐2P  specifications  but  with  the  addition  of  a  distillate  to  slow  the  break  slightly.  

• Most  suppliers  will  offer  the  specification  they  are  producing  to.

Standard  Chipseal

• One  layer  thick  • Asphalt  Residue  glues  the  chips  down    

– It’s  the  main  source  of  strength  for  the  seal        • CRS-­‐2P  needs  clean  chip  to  adhere  to  due  to  fast  break  

– Major  cause  of  seal  failure  is  dirty  chip  • Need  enough  Glue  to  hold  the  chip  

– Major  cause  of  seal  failure  is  not  enough  glue.  

• Need  enough  compaction  before  winter  – Major  cause  of  seal  failure  is  low  compaction  /  rock  is  sitting  up  high

There  is  more  to  a  successful  Design  than  the  numbers

• Timing:  • 160  hours  pavement  temp  exceeding  110    F  

– The  warmth  is  needed  to  soften  the  binder  residue  so  that  the  final  few  percent  of  water  that  is  trapped  in  the  seal  can  work  its  way  out    

– This  means  even  with  a  good  design  if  the  seal  is  done  late  in  the  season  the  final  product  will  not  be  achieved  until  the  next  season.  

– The  earlier  the  better!!    Let  the  warm  weather  help  traffic  finish  the  seal

Holes  left  by  late  water  vapor  escaping  the  seal

There  is  more  to  a  successful  Design  than  the  numbers

• When  the  rock  is  dropped  into  the  binder  the  voids  will  approach  50%  

• Rolling  will  drop  that  to  around  30%  • The  final  product  to  be  achieved  in  the  design  will  not  occur  until  the  voids  are  down  around  20%  

• If  you  don’t  have  enough  traffic,  the  voids  won’t  decrease  and  the  binder  will  not  reach  the  design  level.    Possible  seal  failure  until  final  embedment  /  voids  are  reached.  

Need  more  rolling  

• While  the  wheel  paths  get  the  compaction  needed  for  a  successful  McLeod  designed  seal    

• Parking  lanes,  Fog  line,  Turn  lanes,  Qtr  Crown,  Center  Line  area,  etc.  do  not.  

• Traffic  will  take  much  longer  to  give  these  areas  the  compaction  needed  per  the  design.  

• Give  it  extra  attention  while  building  (Give  them  extra  Rolling)      

• Add  a  STEEL  roller

38,000  lb.  VibratoryBonner  County,  Idaho

Note  the  Effect  on  the  Aggregate  Texture  

½  In.  Chipseal

Maintenance  Seal High  Float  Emulsion  and  Crushed  Cover  

Stone• Use  HF-­‐150  and  Crushed  Cover  Stone  vs.  ½  to  ¼  and  CRS-­‐2P  

• Application  rate  is  .40  to  .46  gal/yd2    vs.  .55+  gal/yd2  for  CRS-­‐2P.      

• This  technology  similar  to  that  used  in  Scandinavian  Countries  “Otta”  Seal  

• Highly  reliable  seals  • TRB  report  No.  1989  Discusses  “Otta”  seal

Maintenance  /  Otta  Seal

• Matrix  of  rock  like  hotmix  – Rocks  interlock  for  strength  

• The  more  fracture  the  better  – Asphalt  Residue  fills  small  voids  and  surrounds  rocks  like  in  hotmix  

– Strength  comes  from  interlock  &  glue  • Twice  the  forces  ,  Adds  to  reliability

 High  Floats  Emulsions Ex:  HF-­‐150

• Slower  setting  than  CRS-­‐2P,  CMS-­‐2P    • Chemistry  reduces  temperature  and  bleeding  susceptibility  

• Very  versatile,  can  be  used  with  very  dirty  aggregate  or  slow  application  processes  

• Soft  residue  allows  traffic  to  knead  the  seal  for  a  longer  time  adding  to  embedment  

• Lower  cost  than  Polymer  emulsions

Asphalt Emulsion HFE-150 Specifications

Test on Emulsion Minimum Maximum

- Viscosity @ 122 Degree F SFS T59 50 400

- Sieve test %, T59 -- 0.3

Distillation:

- Oil Distillate by volume of emulsion % T59 -- 5

- Residue % T59 65 --

Tests on the Residue from Distillation:

- Penetration 77 Degrees F, 100g, 5s T49 150 300

- Float at 140 Degrees F., sec., 1200 --

WSDOT  “crushed  Cover  Stone”    9-­‐03.4(2)

• ¾”  square       100  %  passing  • 5/8”  square       95-­‐100  • U.S.  No.  4       20-­‐45  • U.S.  No.  200     0-­‐7.5  • %  fracture  by  wt.,  min.   75  • Sand  Equivalent  min.     40  • Static  Stripping  test     Pass  

• Much  dirtier  than  chips  used  with  cationic  emulsions  • Much  lower  cost  material  as  we  keep  many  of  the  agg.  Sizes  

vs.  screening  them  off  as  waste  during  chip  production.

Adams  County  Aggregate  SpecCoverstone Maintenance shall meet all the requirements of Section 9 -03.4(2) except 1 that it shall meet the following specifications for grading, fracture and sand equivalent: 2

3 Sieve Size Percent Passing Tolerance Limits 4 3/4" square 100 95-100 5 5/8" square 95-100 90-100 6 US No. 4 20-45 16-49 7 US No. 200 0-5.0 0-6.5 8 % fracture, 9 by weight, min 90 85 10 Sand equivalent min. 40 35 11

12 The third paragraph of Section 9-03.4(2) is revised to read: 13 14

The fracture requirement shall be at least two fractured faces and will apply to the 15 combined aggregate retained on the U.S. No. 4 sieve in accordance with FOP for 16 AASHTO T 335. 17

WSDOT  Crushed  Cover  Stone

Benefits  of  HF  /  Maintenance  Seals

• Lower  EM  use,  lower  rock  cost  =  Substantial  Savings  $$  

• More  forgiving  during  construction  – Slower  to  break,  more  time  to  work  – Handles  dirtier  rock  – Very  low  seal  failure  rate,  always  something  left  

• More  flexible  due  to  softer  residue  – Retard  crack  reflection?

Freshly  broomed  and  washed  Maintenance  seal

Maintenance  Seal  

Tips  for  Successful  Maintenance    Seals

• Maintenance  Seals  create  a  matrix  much  like  hotmix    – Treat  them  like  hotmix  – Higher  traffic  designs  need  to  be  compacted  more  

• Need  to  increase  compaction  while  the  emulsion  is  still  wet  and  can  grab  loose  rock.      – Increased  rolling  effort  means  more  compaction  early,  grabs  rock  before  it  can  slough  off.    

• Increases  thickness  of  seal  (holding  more  rock)  at  equivalent  emulsion  rate.      

Tips  for  Successful  Maintenance    Seals

• Water  can  be  sprayed  over  the  seal  surface  while  rolling  to  help  emulsion  travel  through  the  aggregate  and  expand  coated  surfaces.  

• Higher    traffic  =  more  compaction,  less  room  for  asphalt  residue.  (  Just  like  Hotmix)  – If  you  don’t  compact  enough  during  construction    traffic  will  post  compact  and  flush  the  surface  just  like  under  compacted  hotmix.  

FA-­‐2  Aggregate  SealCity  of  Spokane

• Fine  graded  Seal  using  #4  Agg,  CRS-­‐2P  and  a  post  fog  

• Provides  fine  surface  for  pedestrians  and  local  traffic  

• Seems  tough  –  3rd    yr  seals  holding  up  very  well  • Aesthetically  very  appealing  • Improves  older  surfaces  much  like  slurry

FA-­‐2  Chip  Seal  Specs

• CRS-­‐2P  .15  to  .25  G  /  SY  – Application  at  the  higher  end  holds  multi  layers  of  rock  for  finer  surface  more  surface  correction  

• FA-­‐2  Rock  15  to  25  lbs  /  SY  – Will  sweep  off  a  good  percentage  but  need  to  place  initially  to  absorb  /  account  for  all  the  CRS-­‐2P  

• Rapid  Curing  Fog  Emulsion  .1  to  .15  G  /  SY  – Extra  Insurance  and  gives  a  final  appearance  like  pavement

Issues  with  FA-­‐2  Seals

• Finer  seals  show  deformities  more  than  larger  seals  – Need  to  not  leave  large  bands  of  crackfiller  they  transfer  through  

– Drilling  of  the  seal  can  be  an  issue  with  thicker  emulsions,  smaller  rock  and  this  high  emulsion  rate  will  show  this  more

Points  to  remember• Do  a  design  /  Understand  the  size  and  cleanliness  of  your  rock,  the  condition  of  the  road  and  traffic  count.    

• Do  your  sealing  early  in  the  season  (You  need  cure  time)  – Need  time  to  get  rid  of  the  last  water  – Need  time  to  get  the  rock  compacted  to  its  most  stable  dimension  

• Add  that  steel  roller  and  roll,  roll,  roll,  • If  good  chip  is  hard  to  find,  if  you  can’t  get  the  precision  you  need  for  McLeod,    – Try  a  Maintenance  Seal  


Recommended